Stepping, rather than jumping, towards sustainability

posted in: Science | 4

The director of the WSU Center for Sustaining Agriculture and Natural Resources, Chad Kruger, posed the question, “For farm and food system sustainability, is it possible to shift our existing system in the right direction with small, positive changes or do we actually need to completely redesign our farm and food systems?” I think, with few exceptions, that small changes will be most successful.

In sustainable agriculture circles, and especially among agroecology proponents, it is asserted that a well-designed farming system will encourage self-regulating populations of pests (called homeostatis) while sustaining yields at acceptable levels. Furthermore, this state of self-regulation is claimed to be a property of the system as a whole (system-level emergent). Therefore, when pests do more damage than acceptable, the system is assumed to be wrong; lacking in diversity, using the wrong inputs, too much of this or too little of that, or the system has not been in place long enough to allow these properties to emerge. This assumption has led to the call for more systems-level research, which will lead to “transformational pathways” forward.

Regardless of these claims, I contend that by focusing on such systems our Extension efforts will result in much slower adoption than would a focus on incremental changes. This has to do with the way farmers adopt innovations, the pace of change in agriculture, and the complexity of farming systems.

For a farming system to be transformational, a significant number of farmers must adopt it. However, the nature of farming systems assures that their adoption, under most circumstances, will be slow. In the 2003 book Diffusion of Innovations, University of New Mexico professor and author Everett Rogers states, “Innovations that are perceived by individuals as having greater relative advantage, compatibility, trialability, and observability and less complexity will be adopted more rapidly than other innovations.” Farming systems are complex, and therefore difficult to comprehend in their entirety. They are not amenable to experimenting with on a limited basis because they require the full system to be implemented to gain the presumed systems-emergent properties. In addition, they often do not produce results that neighbors can observe directly. This may explain why the extensive set of integrated farming systems trials established in Europe over the 1980 to 1990 period (2, 3) have had only modest adoption by farmers.

Systems proponents often insist on long-term research. Although I agree that this may be needed to see the long-term effects of systems, it also ignores the pace of change in agriculture. While not the exponential change in computer technology (but definitely affected by it), change in agriculture is happening at a rate that outpaces any attempt at developing farming systems. Market forces, annual weather variations, and dealing with the “pest-of-the-day” will quickly make obsolete any system addressing today’s problems. The reality of these broader, rapid changes, require us to focus on mainly incremental changes on the farm.

Finally, many agroecologists and systems proponents understate the complexity in designing farming systems. Systems theory suggests that there is a restrictive limit to the number of interactions and the level of complexity that can be analyzed. Moreover, the results of farming systems with very specific components cannot be generalized (1, 4). As an example of this complexity, consider the number of variables in the mustard green manure “system” that I have worked with for over ten years: species and variety selection, planting methods and timing, previous crop, management of the previous crop’s residue, soil fertility, incorporation methods and timing, the pest and beneficial community in the target field’s soil, and the ever-present, but highly variable weather factor. Although most would consider this an incremental change, the number of variables in this simple example is large, but much less than the number in a full farming system. I believe that this complexity limits the utility of designed agroecosystems.

In coming to this conclusion that we must rely on incremental changes, I still believe that we can move towards sustainability. If the incremental changes are thoughtful, guided by specific principles, they can take us in the direction we want to go. Another probable scenario, where transformational change is more likely, is change forced upon farmers by disruptive events beyond our control. These will happen, but between them, incremental change will be the rule.


  1. Boulding, K.E. 1956. General Systems Theory–The Skeleton of Science. Management Science, 2:197-208.
  2. Vereijken, P. 1999. Manual for prototyping integrated and ecological arable farming systems (I/EAFS) in interaction with pilot farms.
  3. Vereijken, P. 1992. A methodic way to more sustainable farming systems. Neth. J. of Agric. Sci. 40:209-223.
  4. Weinberg, G. M. 1975. An Introduction to General Systems Thinking. Wiley & Sons, New York.

Many of these references and ideas are found in a poster presentation by Kaffka and Foin. 2006. Complexity and Research on the Sustainability of Farming Systems. ASA-CSSA-SSSA International Meetings, Indianapolis, ID. Abstract

Follow Andrew McGuire:

Andrew McGuire is an Irrigated Cropping Systems Agronomist for Washington State University Extension. He works with farmers in the Columbia Basin of Central Washington in improving soils through cover crops and high residue farming systems.